检索结果-内蒙古大学图书馆

8th International Conference on Information technology, Information Systems and Electrical Engineering, ICITISEE 2024

作者： Deliansyah, Heru Gumilang, Zaky Satya Alfayed, Muhammad Iswanto, Irene Anindaputri Hidayat, Muhammad Fadlan School of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9798350368970

The proliferation of fake news on social media platforms has become a significant societal challenge, undermining public discourse and democratic processes. This study proposes a novel approach to fake news detection by leveraging swarm intelligence techniques, specifically a Modified Ant Colony Optimization (ACO) algorithm, in combination with established machine learning models. Our method employs a four-stage process: data preparation, feature extraction using TF-IDF vectorization, feature selection via Modified ACO, and model evaluation. We compare the performance of Random Forest, XGBoost, and Support Vector Classifier (SVC) models before and after dimensionality reduction. The WELFake dataset, comprising 72,134 news articles, is used to train and evaluate the models. Results demonstrate that the proposed approach achieves high accuracy in fake news detection while significantly reducing computational time, particularly after dimensionality reduction. This research contributes to the development of more efficient and accurate fake news detection systems, addressing the critical need for robust tools to combat misinformation in the digital age. © 2024 IEEE.

关键词： Dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Comparative Study of GPU Performance and Energy Efficiency Across Generational Architectures: A Systematic Literature

Comparative Study of GPU Performance and Energy Efficiency A...

引用

2024 International Conference on Control and Automation, Electronics, Robotics, Internet of Things, and Artificial Intelligence, CERIA 2024

作者： Muhammad, Risang Faiz Syahputra, Muhammad Edo School of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9798331511166

In this study, the evolution of GPU architectures across multiple generations is explored, focusing on their computational performance and energy efficiency. Using a systematic literature review and comparative analysis, this study examines key advancements in GPU technology, particularly in AI workloads, ray tracing, gaming, and scientific simulations. The analysis reveals that modern GPUs, such as those based on the Ampere architecture, demonstrate significant improvements in both performance and energy efficiency, with a 30% increase in AI task performance and a 20% reduction in power consumption compared to previous generations. This study also highlights how innovations in specialized cores and optimized memory management have enabled GPUs to excel in domain-specific applications. Additionally, the broader implications of these advancements for other computational fields, including gaming and scientific research, are addressed. By providing a comprehensive view of these developments, this review offers valuable insights into future GPU design trends and the ongoing challenges in balancing high performance with energy efficiency. © 2024 IEEE.

关键词： Memory management

来源：评论

学校读者我要写书评

暂无评论

Molecule World: Enhancing chemistry education through web-based augmented reality using Assemblr 9th

Molecule World: Enhancing chemistry education through web-ba...

引用

9th International Conference on computer science and Computational Intelligence, ICCSCI 2024

作者： Handoyo, Kevin Joseph Wisnuwardana, Christian Jodi Austen, Aaron Permana, Frihandhika Computer Science Department School of Computer Science Bina Nusantara University Jakarta11480 Indonesia

Our research explores the role of web-based AR, namely Assemblr, in creating an interactive, engaging, and immersive learning environment. The focus of this research is how the use of AR can motivate, engage, and increase students' interest in learning in the field of education, especially in the field of chemistry on molecules and compounds. The general impression of the Molecule World app, built within the ADDIE instructional design framework, significantly improved the user experience. The attractiveness of the app achieved a benchmark score of 1.88, dependability a benchmark score of 1.73, and novelty a benchmark score of 1.65. Perspicuity can be improved with a benchmark score of 2.00, efficiency of 1.80, and stimulation of 1.63. The results of our research show that augmented reality technology can be an alternative for the world of education, especially in the field of chemistry, to increase the attractiveness and interest of student learning. We hope that this research can open new insights related to the application of immersive technology, namely augmented reality, which is implemented into the world of education to improve the quality of chemistry learning. © 2024 The Authors.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Color-Based Auto-Rickshaw Detection Using Convolutional Neural Networks for Intelligent Transportation Systems 2

Color-Based Auto-Rickshaw Detection Using Convolutional Neur...

引用

2nd International Conference on Information and Communication technology, ICICT 2024

作者： Hasan, Md. Mehedi Aman, Nakib Jahan, Md Wadud Developer Experience Hub Rajshahi Bangladesh Department of Computer Science and Engineering Pabna University of Science and Technology Pabna Bangladesh Department of Computer Science and Engineering Varendra University Rajshahi Bangladesh

ISBN: (纸本)9798331508227

The investigation in this paper is centered on a new methodology for improving the effectiveness of intelligent transportation systems by identifying and distinguishing autorickshaws using Convolutional Neural Networks (CNN). The strategy proposed involves training a CNN model to automatically recognize and differentiate the colors of auto-rickshaws, which in turn significantly enhances the detection capabilities. Using hierarchical convolution feature models, the CNN algorithm independently extracts relevant features from images, thereby aiding in precise object detection. An assessment of the model's performance is conducted using a carefully selected dataset containing 1000 auto-rickshaw images, illustrating a high level of accuracy with minimal margin of error. This research makes a valuable contribution to the advancement of automated transport detection within the realm of traffic management. It presents a more streamlined and impactful methodology aimed at improving the overall efficiency and reliability of the system. ©2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

TL-IDS: A Transfer Learning Technique for Botnet Detection in IoT 6

TL-IDS: A Transfer Learning Technique for Botnet Detection i...

引用

6th International Conference on Computing and Informatics, ICCI 2024

作者： Abdelhamid, Salma Hegazy, Islam Aref, Mostafa Roushdy, Mohamed Computer Science Department Faculty of Computers and Information Technology Future University in Egypt Cairo Egypt Computer Science Department Faculty of Computer and Information Sciences Ain Shams University Cairo Egypt

ISBN: (纸本)9798350373875

With the exponential rise of Internet of Things (IoT) devices in our modern-day lifestyle, the potential danger of botnet intrusions has become inevitable. These botnet attacks cause extensive damage to both individual users and large organizations. In the realm of IoT botnet intrusion detection, learning-based Intrusion Detection Systems (IDS) were proposed as a powerful strategy for identifying malicious behaviors and consolidating the security of IoT systems. Moreover, Transfer Learning (TL) techniques have emerged to overcome the problems of scarce data and the long training time of traditional learning approaches. This paper proposes an IDS for IoT networks based on the TL EfficientNet model. We train the system and validate it using the Bot-IoT network intrusion dataset. Our approach incorporates balancing the dataset, redundancy removal, dimensionality reduction, and transformation of the dataset into RGB images to be fed as input to the detection model. The findings of our experimental work prove that the proposed model is robust and effective, and it surpasses existing models with an accuracy of 99.53%, recall of 99.53%, and F1-score of 99.52%. © 2024 IEEE.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

TWISH: an enhanced activation function for optimizing mental health and comorbidity risk assessment during COVID-19 using deep learning

引用

Multimedia Tools and Applications 2025年 1-21页

作者： Bhimavarapu, Usharani Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Andhra Pradesh Vaddeswaram India

Significant health hazards have been associated with the COVID-19 pandemic, especially for women, whose mental and physical health have suffered greatly. In order to predict the risk of COVID-19 in women, this study investigates the use of stacked artificial neural networks (ANNs) in hidden layers with a proposed Twish activation function. The study incorporates many psychological and health-related measures into the artificial neural network framework, facilitating an all-encompassing evaluation of risk. Important measures include the Beck Anxiety Inventory (BAI), which assesses anxiety specifically related to pandemics (ANN1), the Social Readjustment Rating Scale (SRRS), which assesses stress from significant life events (ANN2), the Centre for Epidemiologic Studies Depression Scale (CES-D), which assesses symptoms of depression (ANN3), the Charlson Comorbidity Index (CCI), which assesses the impact of pre-existing health conditions (ANN4), and the Prenatal Distress Questionnaire (PDQ), which measures stress in pregnant women during the pandemic (ANN5).The ANN model enhances performance in the hidden layers by using the Twish activation function, enabling more precise and effective COVID-19 risk prediction. Predicting risk variables for women is especially helpful because of the intricacy of mental health, co-occurring conditions, and stress associated with pregnancy during the pandemic. All classes reveal that the proposed model is remarkably accurate, and several of its modifications produce outstanding outcomes. With a 97.75% accuracy rate, ANN1 has the strongest predicting capability. With accuracy scores of 97.50%, ANN2 and ANN4 continuously demonstrate good performance, highlighting dependable performance. ANN3, which has an accuracy of 98.00%, draws attention to outstanding performance. Furthermore, with an accuracy rate of 97.50%, ANN5 retains great predictive capabilities, demonstrating good and consistent performance in a variety of circumstances. © The A

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Optimizing Pretrained ResNet Models for Alzheimer's Disease Detection in MRI Images 7

Optimizing Pretrained ResNet Models for Alzheimer's Disease ...

引用

7th International Conference on Informatics and Computational sciences, ICICoS 2024

作者： Prasetyo, Simeon Yuda Bina Nusantara University School of Computer Science Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9798350375886

Alzheimer's disease (AD) is a serious concern in healthcare because it affects cognitive skills and quality of life. Recent advances in medical imaging and machine learning techniques provide great opportunities for early identification and diagnosis of Alzheimer's disease. This study explores the optimization of pretrained ResNet models for AD detection in MRI images, focusing on ResNet50, ResNet101, and ResNet152 architectures. The research comprises two phases: transfer learning and fine-tuning. Results show that deeper model architectures perform better in the transfer learning phase, with ResNet152 demonstrating superior performance, achieving an accuracy of 99.417%. Fine-tuning significantly enhances model accuracy, with all models achieving nearly perfect scores. ResNet152 emerges as the top-performing model, emphasizing the importance of fine-tuning in optimizing classification accuracy for AD detection. Future study might concentrate on fine-tuning model architectures and using sophisticated regularization approaches to improve AD classification. This work helps to the development of accurate diagnostic methods for early Alzheimer's disease identification, allowing for prompt intervention and treatment techniques. © 2024 IEEE.

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

A Data-aware Probabilistic Client Sampling Scheme in Streaming Federated Learning

A Data-aware Probabilistic Client Sampling Scheme in Streami...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Song, Chao Huang, Jianfeng Wu, Jie Lu, Li University of Electronic Science and Technology of China School of Computer Science and Engineering China Temple University Department of Computer and Information Sciences United States

ISBN: (纸本)9798350351255

In streaming federated learning, where data on each client is received in the form of a data stream, the distribution of data on the clients has a significant impact on the performance of the federated learning model. The continuous influx of streaming data on the clients leads to real-time changes in the local data distribution, which in turn affects the performance of the federated learning model. Furthermore, the heterogeneity in data distribution among clients exacerbates this impact. In this paper, to address these challenges, we propose a Data-aware Probabilistic Client Sampling scheme (DPCS) for selecting appropriate clients to participate in model training in each round of federated learning. DPCS begins with a method for real-time monitoring of local data distributions on the clients. Based on these observations, the central server adopts a probability-based client sampling strategy. Through extensive experimentation, we demonstrate that our client sampling scheme offers higher timeliness and enhances the performance of federated learning compared to traditional methods. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Refresh Learning for Dynamically Changing Fresh Produce Classification in Continual Learning Scenario

Refresh Learning for Dynamically Changing Fresh Produce Clas...

引用

2024 IEEE International Conference on Internet of Things and Intelligence Systems, IoTaIS 2024

作者： Vincencius, Bryan Felice, Eldwin Reidolvo Jabar, Bakti Amirul Kurniadi, Felix Indra Computer Science Department School of Computer Science Bina Nusantara University Jakarta11530 Indonesia

ISBN: (纸本)9798331521219

Precise and prompt product identification is vital for efficient inventory management in the retail grocery industry. While deep learning models have significantly improved in classification accuracy, dynamically changing environments, seasonal variations, and new products pose significant challenges. These factors create new data on which pre-trained models are not adequately trained, leading to a substantial drop in accuracy. Continual learning has emerged as a solution, allowing models to adapt to new data without forgetting previously learned information. However, this solution still suffers from catastrophic forgetting, especially when dealing with large datasets. This paper aims to explore strategies to mitigate catastrophic forgetting and develop a robust model for fruit and vegetable classification in the retail sector. The models used are two memory-based models (DER++ and ER) and two regularization-based models (EWC and EWC + CPR). The result indicates that memory-based models outperform regularization-based models, with Dark Experience Replay++ (DER++) showing the highest accuracy in both class-incremental and task-incremental settings. Memory-based models are less affected by catastrophic forgetting as they have direct access to past data, allowing them to have better knowledge preservation. Furthermore, the study also reveals that hybrid approaches (e.g., EWC + CPR) can further enhance retention accuracy. ©2024 IEEE.

关键词： catastrophic forgetting continual learning image classification memory buffer regularization

来源：评论

学校读者我要写书评

暂无评论

AFMA-Track: Adaptive Fusion of Motion and Appearance for Robust Multi-object Tracking 27th

AFMA-Track: Adaptive Fusion of Motion and Appearance for ...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Liao, Wei Luo, Lei Zhang, Chunyuan College of Computer Science and Technology National University of Defence Technology Changsha China Science and Technology on Parallel and Distributed Processing Laboratory College of Computer Science and Technology National University of Defense Technology Changsha China

ISBN: (纸本)9783031784439

Motion and appearance cues play a crucial role in Multi-object Tracking (MOT) algorithms for associating objects across consecutive frames. While most MOT methods prioritize accurate motion modeling and distinctive appearance representations, the use of appearance and motion cues is often confined to simplistic association techniques. For instance, fixed weights are commonly employed to combine the intersection-over-union (IoU) matrix and appearance similarity matrix, yielding an association cost matrix. To harness the full potential of motion and appearance cues across diverse scenarios, we propose an innovative approach that dynamically balances motion and appearance cues based on scene and object information during the association process. Furthermore, we introduce a new mechanism for updating appearance representations, effectively mitigating noise introduced by occlusion. Our method demonstrates state-of-the-art performance on the MOT17 and MOT20 test sets. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：